Semantic Sensing Performance Analysis: Assessing Keyword Coverage in Text Data

doi:10.1109/TVT.2023.3282742

Semantic Sensing Performance Analysis: Assessing Keyword Coverage in Text Data

2023-01-01

7

PRE

AI

DOI：10.1109/TVT.2023.3282742

Original

Original request for help

Save

Abstract

En 中文

In this paper, to provide the theoretical reference for semantic sensing performance, the stochastic geometry tool is used to make the quantitative coverage performance analysis of the semantic keywords. This method can help understand semantic network performance from the macroscopic view rather than the connection-oriented perspective. Firstly, through the semantic mapping operation, the context space is transformed into the semantic space in the semantic sensing process. Secondly, given the text data with different statement styles, e.g., literary and scientific texts, two typical semantic reflection models are established with the help of stochastic geometry theory, where the keyword coverage expressions are derived. Thirdly, under various parameter settings, the correctness and effectiveness of the proposed models have been evaluated through the simulation results and analysis.

Keywords:

Semantics

Reflection

Sensors

Stochastic processes

Geometry

Performance analysis

Channel models

Semantic sensing

stochastic geometry

keyword coverage performance

semantic communication

Semantic Sensing Performance Analysis: Assessing Keyword Coverage in Text Data

Abstract

Journal

Organization